Spectrometric Small Pixel X ray Imager – SPEPIX

Development of a small surface readout test circuit, which main objective is to test the analog performance of the detection chain.
From the expected performance of the test circuit, development of the output data process, and simulation of expected output images.

The test circuit has been realized and is under characterization (see the results in corresponding publications).
Data processing is based on image material base decomposition.

Improvement of material based image decomposition, realistic simulations related to the medical needs.
To improve energy resolution, development of a second version of the test circuit, including in particular pixel binning in order to collect the charges corresponding to a single detected X-ray which are spread out over several pixels.

Communications:
- IEEE-NSS-MIC conference, 2013 : «Sphinx1: Spectrometric Photon Counting and Integration Pixel for X-ray Imaging with a 100 electron LSB « (A. Habib, M. Arques, B. Dupont, P. Rohr, G. Sicard, M. Tchagaspanian, L. Verger)
- IEEE-NSS-MIC conference, 2014 : Characterization Results of Sphinx1 ASIC for X-Ray Detection by Photon Counting and Charge Integration with a 100 Electron LSB (A.Habib, M. Arques, B. Dupont, P. Rohr, G. Sicard, M. Tchagaspanian, L. Verger)

Publications:
- Light yield sensitization by X-ray irradiation of the BaAl4O7:Eu2+ceramic scintillator obtained by full crystallization of glass DOI (G. Patton, F. Moretti, A. Belsky, K. Al Saghir, S. Chenu, G. Matzen, M. Allix, C. Dujardin): Phys. Chem. Chem. Phys., 2014,16, 24824-2482910.1039/c4cp04064c
- to be published in TNS (Transactions on Nuclear Science) :
Sphinx1: Spectrometric Photon Counting and Integration Pixel for X-ray Imaging with a 100 electron LSB (A.Habib, M. Arques, B. Dupont, P. Rohr, G. Sicard, M. Tchagaspanian, L. Verger)

Flat panel radiological detectors, such as those sold by TRIXELL company, today mostly replace X-ray films (for single image imaging) and X ray Image Intensifiers (for video imaging). They offer for these applications, a large surface from 20cm x 20cm up to 43cm x 43cm, a good spatial resolution with pixels ranging from 100µm x 100µm to 200µm x 200µm, and a good temporal resolution with video rates of 100 images per second or more.

Besides, the interest for distinguishing X-ray photons energy has been identified for a long while. This identification, or spectrometry, also called "color imaging", allows separating patient constituents versus their chemical composition. With a bi-energy detector, bones, which absorb both low energy and high energy X rays, can be separated from soft tissues which are mostly transparent at high energies. Any additional energy channel (three-energy, four energy,... detector) helps bringing more information.

Spectrometric detector research has brought to market X ray scanner equipments for major companies (Siemens, General Electric, Philips).
The goal of this project is to allow flat panel detectors to also access to the spectrometric imaging performance, thus giving them a very innovative characteristics and a key competitive advantage.

This project deals with the two ends of the problem.
On one end, the design of the detector will be done, specially focusing on the X ray to visible photon converting layer, and on the readout matrix. This design will be followed by the foundry and the characterization of a small surface test circuit.
On the other end, large surface images will be done, but with too much costly detectors (from industrial point of view). These images will allow, with test objects or anatomical phantoms, for the evaluation of the clinical interest of spectrometric detectors. They will also help defining a more precise detector specification.

To reach these goals, the project associates: TRIXELL, industrial world leader of radiological flat panel detectors; CEA-LETI, with its knowledge in innovative X-ray systems design and its facilities in design and characterization of integrated circuits; LPCML (Physical and Chemical Laboratory for Luminescent Materials); and CREATIS, linked to the Lyon-Bron Louis Pradel Cardio-vascular and Respirology Hospital, for their ability to evaluate the clinical interest of the images and to develop software improvement methods.